The concept of social bases of self-respect distinguishes Rawls's framework from purely materialist theories of justice. A society that provided ample income, wealth, and formal rights but systematically undermined the conditions under which citizens could see themselves as contributing members of the cooperative enterprise would fail Rawls's standard even if every measurable resource were distributed justly. The recognition that self-respect requires social conditions — not merely private psychological states — is one of the most philosophically consequential features of the theory.
Applied to the AI transition, the framework illuminates a dimension of the transition's costs that purely economic analyses tend to miss. When AI tools can approximate the outputs of experienced practitioners — when the senior engineer's embodied expertise is matched by a junior colleague with a subscription, when the writer's craft is reproduced by a model trained on centuries of literature, when the teacher's knowledge is accessible to any student with a prompt — these practitioners may retain employment and income, but the social recognition that underwrote their sense of worth is eroded. The erosion is real. It manifests in depression, disengagement, the quiet withdrawal that elegists describe when they speak of mourning something they cannot quite name.
The institutional implications are significant. Protecting the social bases of self-respect requires more than providing transitional income or retraining infrastructure. It requires institutional recognition of the continuing value of human judgment, craft, and accumulated experience — the dimensions of contribution that AI cannot replicate but that the market may nevertheless fail to reward adequately. This might take the form of professional development pathways that valorize the judgment and integrative capability that experience produces; certification and recognition systems that make this value legible; workplace structures that create visible roles for mentorship, quality assessment, and architectural vision that seasoned practitioners provide. The specific mechanisms are for the legislative stage. What the Rawlsian framework provides is the principle: the social bases of self-respect must be protected, because without them, the liberty that the first principle guarantees is hollow and the capabilities that fair equality of opportunity promises are ornamental.
Axel Honneth's recognition theory, though developed independently of Rawls, supplies empirical and phenomenological depth to the concept. Honneth's framework identifies three forms of recognition — love, rights, and social esteem — each producing a distinct dimension of selfhood. The skill devaluation injury that AI inflicts on practitioners whose expertise has been commoditized maps directly onto the third form of recognition. The moral injury Honneth describes is the specific harm that occurs when a social order violates the legitimate expectations of reciprocity underlying its recognition structure.
Rawls introduced the social bases of self-respect in A Theory of Justice (§67) and developed the concept across subsequent work. Its priority among the primary goods grew across his career as he came to see self-respect as the condition on which the value of all other goods depends.
Priority among primary goods. Self-respect is the primary good without which others lose their value; protecting its social bases is a primary requirement of justice.
Institutional, not merely psychological. Self-respect requires social conditions — recognition, meaningful activity, associational membership — not merely private self-regard.
Erosion through commoditization. When AI tools approximate the outputs of experienced practitioners, the social recognition underwriting practitioners' self-respect is eroded regardless of whether their employment is preserved.
Connection to recognition theory. Honneth's framework provides empirical and phenomenological depth to the concept, identifying specific forms of recognition whose denial produces specific forms of moral injury.
Institutional protection required. Justice requires institutions that create visible roles for the dimensions of contribution AI cannot replicate, not merely transitional income for those whose previous roles have been commoditized.